Interactive Innovation of Technology for Mobile Work

Interactive innovation of technology for

mobile work

Jan Kietzmann

Simon Fraser University, Burnaby, BC, Canada

Correspondence: Jan Kietzmann, Faculty ofBusiness, Simon Fraser University,250-13450-102nd Avenue, Surrey,British Columbia V3T OA3, Canada.E-mail: [email protected]

Received: 7 August 2007Revised: 4 December 20072nd Revision: 12 June 20083rd Revision: 24 June 2008Accepted: 25 June 2008

AbstractDespite the increasing popularity of mobile information systems, the actual

processes leading to the innovation of mobile technologies remain largely

unexplored. This study uses Action Research to examine the innovation of amobile RFID technology. Working from Activity Theory, it departs from the

prevalent product-oriented view of innovation and treats technology-in-the-

making as a complex activity, made possible through the interaction ofmanufacturers, their organisational clients and their respective mobile workers.

The lens of a normative Interactive Innovation Framework reveals distinctive

interaction problems that bear on the innovation activity. In addition todifficulties emerging from dissimilar motivations for the innovation project, the

mobile setting presents unique contradictions based on the geographical

distribution of its participants, the diverse role of mobile technology, the

complexity of interacting through representations and the importance ofthe discretion with which mobile work activities are carried out today.

European Journal of Information Systems (2008) 17, 305–320.

doi:10.1057/ejis.2008.18

Keywords: interactive innovation; mobility; technology for mobile work; mobiletechnology; action research; activity theory

IntroductionThe past decades have unveiled an impressive trajectory of mobiletechnology, marked primarily by the advancements from early mobiletelephones, available to the elite few, to the highly advanced and diffusedmobile devices of today. Intrigued by new technological affordances(Gibson, 1977), Information Systems (IS) scholars increasingly turn theirattention to the impact of mobile devices on individuals and organisa-tions. In the meanwhile, manufacturers, concerned with staying compe-titive in an industry actively rivalling for corporate accounts, are starting toinvite their future users to participate in the process of innovation.

These evolving cooperative innovation projects between manufacturersand users are highly complex and benefit from an innovation approachthat clearly defines the roles and responsibilities of each participant. Somelessons can be learned from fairly recent co-constructive developments;however, these lessons are largely based on single organisational contexts,non-mobile computing devices and a dialogue between innovators andusers in fixed-location settings. But since mobile settings and technologiesthat span time and location fundamentally challenge the interactionhabits subsumed by non-mobile contexts, manufacturers and innovatorssupport a fresh approach for interactive innovation within mobileenvironments.

Commissioned by a leading mobile device manufacturer, our researchfocused on first developing a conceptual framework for interactive

European Journal of Information Systems (2008) 17, 305–320

& 2008 Operational Research Society Ltd. All rights reserved 0960-085X/08

www.palgrave-journals.com/ejis

innovation in mobile settings, and second assessing suchframework in a real innovation project. A fundamentalassumption of this inquiry was that innovation as aninteractive activity in mobile work settings involvescategorically different participants, including innovators,primary and secondary users. The aim of this paper is toanalyse how these groups and their respective workactivities established an influential context for theinteractive innovation of new technologies for mobilework.

We pursued this research objective in a 1-year ActionResearch study. The technology under development was asynchronous, mobile radiofrequency identification, or‘mobile RFID’ device, and the innovation activity wasdriven by the innovator, a handset manufacturer anon-ymised here as Nalle. The company, experienced indeveloping personal mobile devices, understood that itneeded to ground this organisational innovation withinreal mobile work settings. Nalle found an appropriatesponsor for its project at Morrison Patrolling, a securityservices company that had an interest in learning moreabout how their mobile security guards conducted work.

Morrison Patrolling’s problem was that the organisa-tion only had a basic understanding of how its mobilesecurity guards carried out work. Essentially, a mobilesecurity guard arrived at work, collected a car andnavigated an enormous terrain throughout his workdayto secure buildings, check the status of schools, patrol theperimeter of construction areas, etc. Throughout thistime, Morrison had no information about the specificlocation of each guard and the status of the premises.When a guard was needed for an emergency, a suspectedbreak-in for example, all guards who could possibly be inthe vicinity needed to be called via their mobile phones.This uncertainty also affected clients, who never knew iftheir premises had been checked, and therefore calledMorrison’s managers, who then had to contact the guardvia mobile phone. By and large, the information flow toand from those in the field was very inefficient, and asolution to Morrison’s problems was sought.

Morrison and its mobile workforce decided to partici-pate in the interactive innovation activity of mobileRFID, a technology that promised to address many ofthese mobility-based shortcomings. RFID essentiallyconsists of two components. A number of transponders,or tags, in the shape of small stickers, would be attachedto various elements of the premises patrolled byMorrison’s guards (e.g., to a gate). The main innovationwas the mobile transceiver, a synchronous reader thatcommunicated with the tags once they were in closeproximity. Once tag and reader had connected, the guardcould select a message on the reader, for instance ‘all ok’.Immediately, the serial number of the tag, tied to thespecific premise, would be sent along with a time-stampand the selected message to Morrison’s office. Thissynchronous information exchange would also updatethe field data stored on an extranet, so that Morrison’smanagers as well as the company’s clients knew where

the mobile guard had been and what the status of eachpremise was.

In the process of developing this innovation, bothNalle and Morrison relied on the help of Morrison’smobile employees. These mobile workers formed theprimary users for the interactive innovation activity,those who would use the actual artefact under develop-ment. As secondary users of Nalle’s technology (Friedman& Cornford, 1992), Morrison’s managers would not usethe technology directly but would benefit from tyinginformation gathered from the field with the help ofmobile RFID into existing corporate IS.

The empirical findings were analysed using Enges-trom’s interpretation of Activity Theory (AT). Particularlyhis framework of activity systems allowed the researcherto examine interactive innovation as a complex processthat involved different groups of participants. AT’spremise of contradictory motivations played an impor-tant role in the analysis of the individuals’ contributionto the collective activity of developing the handheld,synchronous RFID device. The resulting discussionacknowledges both social and technical constituents ofIS that are tied to mobile work practices; it does notspecifically focus on mobile RFID affordances (Angell &Kietzmann, 2006). To understand how participantsactually interacted, the principle of mediated actionsenabled the researcher to look at the role of tools,including mobile technologies, and explicit and implicitrules that affected the activity and its outcome.

This study yielded a number of interesting results.Specific to the scope of the project, it led to manyrefinements of the mobile RFID systems withinMorrison’s organisational settings. Beyond the study, ithelped Nalle innovate its mobile RFID technology for amuch wider audience. More conceptually, the study ledto the development of an Interactive Innovation Frame-work that was evaluated empirically during the study.Here, the theoretical result of its analysis quite clearlyunderlined that interactive innovation is different whenconducted in mobile environments. The very specificcomplexity of the activity of innovation, in AT parlance,resulted from unique contradictions of motives of itsparticipants, from the effects of geographical distributionand modalities of mobility and from contradictoryroles that technology assumed. Further contradictionsemerged from representations as boundary-spanningobjects and contradictory interaction requirements ofthe discretionary practices of mobile workers and theinteractive, knowledge-sharing innovation activity.

The following section briefly introduces aspects ofmobility and innovation that are relevant to this study.Next, AT with its focus on motivation, object-oriented-ness and transformation is introduced as a theoreticalfoundation. Contradictions, as the impetus for changewithin these activities, are emphasised as the mainanalytical tool. After setting the theoretical foundation,the individual stages of Action Research provide thestructure for the subsequent sections of this paper. The

Interactive innovation of technology for mobile work Jan Kietzmann306

European Journal of Information Systems

diagnosis outlines the motivation for the interactiveinnovation project, the planning stage provide details ofthe normative interaction framework that was developedfor it, the action taking stage describes the introductionof the mobile RFID technology, and the evaluation andlearning stages outline the results, for scope of the projectand more generally for interactive innovation of tech-nology for mobile work.

The messiness of mobile work and innovation oftechnologyIn the midst of recent advances of mobile technology andemergence of new mobile devices, it is hardly surprisingthat mobility is at the fore of academic work, too.New developments are theorised as emerging mobile(Sørensen & Pica, 2005), pervasive (Hansmann et al.,2003), nomadic (Lyytinen & Yoo, 2002b) and ubiquitous(Lyytinen et al., 2004) constellations of work andinteraction.

In the course of these conceptualisations, some authorsprimarily study technological aspects of mobility (e.g.,Pierre, 2001; Izadi et al., 2002; Kim et al., 2003), whileothers examine mobility from a social (e.g., Castells,1996; Ling, 1998; Urry, 2000; Agre, 2001; Fortunati, 2001;Plant, 2001) or a socio-technical research perspective(e.g., Kristoffersen et al., 1998; Wiberg & Gronlund, 2000;Lyytinen & Yoo, 2002a; Pedersen & Ling, 2003; Pica &Sørensen, 2004). Contrary to earlier arguments thatproclaimed the proverbial Death of Distance (Cairncross,1997) and Anytime, Anywhere computing (Kleinrock,1996), this abundance of recent mobility-related researchprojects and publications highlight that space, locationand the distribution of data, individuals and devicescontinue to be important factors of human interaction(Olson & Olson, 2000; Wiberg, 2001).

In this context, mobility is associated with theconscious, rational choice of people to move, to meanderand to change location, often in a fluid, unstable way andat times unpredictable to themselves and to others(Kakihara & Sørensen, 2001; Brown & O’Hara, 2002;Lilischkis, 2003). Various modalities of mobility delimitmobile work according to the mode of transportation anduse of technology (i.e., travelling occurs when peoplemove in vehicles), the time spent at any one site (i.e.,visiting occurs when a person temporarily stays at oneplace on a transitory basis before moving on) andlocation (i.e., wandering occurs when a person movesabout a building or specific premises) (Kristoffersen &Ljungberg, 2000). Throughout their workdays, mostmobile workers negotiate a multitude of modalities andcontexts (Scheepers et al., 2006); they may drive or walk,then spend time at a specific place, perhaps wander aboutand then continue to drive to different sites.

At first sight, the affordances of mobile technologysuggest that others could easily relate to such diversemobile work activities from a distance, that mobileinformation systems provide the communication linkbetween those in the field and those in the office.

However, the cooperation, coordination and collabora-tion (Kaptelinin & Nardi, 1997) with mobile workers andthe connection to mobile work continue to present aconsiderable struggle. What truly happens in the field isreally only known to mobile workers. The rich informa-tion from the multitude of contexts they navigate is oftenreduced to paper logs or summarised through voiceconversations with mobile phones. In this process,important contextual detail gets lost, requiring frequentverification and validation between communicationpartners. Similarly, not all mobile work contexts aresuitable for all mobile technologies, and mobile workersselect how to deploy and appropriate technologies incorrespondence to their actual needs in the field. Itappears that mobile technologies do not fully respond tothe needs of mobile work, and at times exacerbate ratherthan improve mobile work and managerial efforts(Wiredu & Sørensen, 2005).

In order to address these difficulties and to improve theinteraction among mobile and non-mobile individuals,manufacturers are busy innovating new devices. Caughtin the traditional trajectory of invention–innovation–diffusion, mobile devices are often developed before usersbecome actively involved. The actual mobile workactivities that these innovations are to serve are black-boxed, and users are only involved in testing thefunctionality and interface design. At this stage, theinnovations are already conceptually fixed, closed, andnon-malleable, and the primary concern of innovationstudies is the technology’s effect on the user. Examplesinclude innovations that are regarded as elicitingdisembodied changes to knowledge and skill sets (Cohen& Levinthal, 1990), leading to competitive advantage(Porter & Millar, 1985; Davis & Devinney, 1997), orintroducing key changes in production or processmanagement in the work organisation (Whipp & Clark,1986; Christensen, 1997; Tidd et al., 1997, p. 23; Porter,2001). In most cases, these effects are examined post-innovation, based on historical accounts and before-and-after comparisons. The innovation practice is treatedpurely as an outcome, the opus operatum, rather than inconcert with the activity of innovation, the opus operandi(Bourdieu, 1977).

Those who involve the user as a valuable input into theinnovation process often pay close attention to howconsumers and lead-users modify existing products.Bricolage, user-innovation and technologies that emergein situ demonstrate how the use of technologies can differfrom their intended application (Ciborra, 2002; Shah &Tripsas, 2004; Haddon et al., 2005; Von Hippel, 2005).Nonetheless, these phenomena still do not describeco-constructive activities or the underlying motivationof the various parties involved. However, they underlinethe disconnect between existing innovations and theactual needs of their users.

With a focus on the causal involvement of the user andthe activity of co-constructive efforts, studies have beenconducted in non-mobile environments. An essential

Interactive innovation of technology for mobile work Jan Kietzmann 307


premise is that developers and users are teams of differentexperts (Bjerknes & Bratteteig, 1987), and that innova-tions ‘should be done with users, neither for them nor bythem’ (Ehn & Kyung, 1987, p. 54). Participatory Design,for instance, underscores the involvement of users in theplanning and designing of information systems (Floydet al., 1989; Bødker & Grønbæk, 1991; Ehn, 1993; Jones,1995, p. 72; Piller et al., 2003). Particularly Bødker’sparticipatory human–computer interface (HCI) designstudies adopt such an activity-lens; however, Participa-tory Design and interactive innovation place the user in afundamentally different role.

Participatory Design, with the underlying assumptionthat design is mostly concerned with determining details tomeet a particular purpose, focuses on an emancipatoryelement that is guided by conflicts and concerns asperceived by the user. Examples include HCI work, inwhich the functional properties of the respective devicesare already established and inscribed into the artefact.Users are involved to identify the drawbacks of currentinterfaces and are tasked with improving their usabilityand usefulness (Bødker, 1991), without a direct focus onthe actual functions, the affordances, of the device.

Interactive innovation, on the other hand, is first andforemost concerned with determining the purpose of aninnovation. In other words, innovators invite users tohelp develop the functions that the device is to serve. Theresulting back-and-forth between innovators and usersindicates that contradictions, perhaps between what istechnically possible and what users demand, requirefurther negotiation. Such a struggle emphasises thatlearning occurs when these two parties try to align theirrespective activities, which then informs the actualinnovation process and the resulting product itself.Overcoming these contradictions during this collectiveinnovation activity affects the actual purpose the in-novation will serve, not just its design component.

Consequently, a practice-based approach to innova-tion, one that involves the interaction of a number ofparticipants, motives and concurrent activities, respondsmost favourably to a mobile reality that is often messy,non-linear and not necessarily sequential. Nonetheless,despite calls for investigating these emerging practice-based activities (Fontana & Sørensen, 2005; Scarbrough &Swan, 2005) within the ‘bewildering array of definitionsand approaches’ regarding innovation (Swan & Newell,2000, p. 27), the causal involvement of the user hashardly been studied. The interactive innovation lensintroduced below treats the process of innovation ‘not ina normative or naturalistic way, but as a sociallyconstructed constellation of activities and practices’(Scarbrough & Swan, 2005, p. 2) involving technologyand mobile work.

Innovation and the theory of activityAn investigation that aims to examine the process ofinteractive innovation, not its outcome or its effect onthe user, requires a theoretical framework centred on

collective activities. We selected AT (Leontiev, 1974,1978; Vygotsky, 1978; Engestrom, 1987) as an analyticallens that focuses on the context of meaningful, goal-oriented and socially determined interaction betweenhuman beings and their material environment (Bannon,1997).

AT lends itself to IS studies in general and to mobilitystudies in particular as it pays particular attention to therole of tools and artefacts as mediators of humanactivities. In this realm, a number of ISscholars (includingBødker, 1991; Nardi, 1995; Wiberg, 2001; Wiredu &Sørensen, 2005) have demonstrated the usefulness of AT.Especially Engestrom’s interpretation and extension ofAT (1987), through his framework of activity systems(Figure 1), has become a recognised conceptual frame-work for describing the structure, development andcontext of computer-supported activities (Kuutti, 1996;Kaptelinin & Nardi, 1997).

In an A T approach to understanding human beha-viour, ‘all human experience is shaped by the tools andsign systems we use’ (Nardi, 1995, p. 5). These so-calledmediators are represented above through tools, rules anda division of labour that mould our interaction withobjects in the real world. In AT, subjects’ activities aremotivated to transform these objects into desired out-comes (Vygotsky, 1978). For instance, a Nalle engineerused his tools, physical or intangible, to transform aregular mobile phone into an RFID-enabled handset. Amobile security guard used his keys to transform anunlocked building into a secured premise. Accordingly,each activity is expressed through a clear object-orient-edness, and the motivation to transform each object intoa new outcome.

Favourable for this study, Engestrom’s activity systemssuggest a move from an instrumentalist approach oflooking at discrete tools, etc. to an interactionistperspective that supports the holistic interplay of theinherent elements. In recognition of the impact of thesupra-individual influences on activities, even when asubject is apparently working alone, the communitycomponent gives weight to the social and culturalcontext of the work environment and to neighbouringactivities. For an interactive innovation study, this focusis of paramount importance, as it allows the researcher to

Tools

Subjects

RulesCommunity

Division of Labour

Object OutcomesTransformation

Process

Figure 1 Engestrom’s framework of Activity Systems (1987).



examine the activities of Nalle, Morrison and the securityguards and their impact on each other’s activities.

In this complex view of activity systems, all theelements are interdependent and (re)shape each otherthroughout the duration of the collective activity.However, the framework of activity systems does notsuggest harmony between all of its elements. On thecontrary, the value of their interaction lies in the learningthat occurs during the negotiation of problems that occureither between the nodes (for instance when a subjectdoes not have the appropriate tools for the activity), orwithin them (for example when there are two rules thatcontradict each other). These problems, called contra-dictions in AT, may at first sight be seen as complicationsfrom an innovation perspective. Nonetheless, they alsolead the activity system through constructive, expansivecycles, in which the negotiation of contradictionsprovides the sources of new knowledge and drives theresulting transformation of an object into an outcome.

When considering a complex collective activity such asinteractive innovation, neighbouring activity systems ofinnovator, primary and secondary users add to thepotential for problems and to the learning that can arisefrom their interaction. The fluid negotiation of contra-dictions within and between their activities permits apractice-oriented view of interactive innovation insteadof a mechanistic product-oriented perspective of innova-tions. In combination with its premises of motivation,mediation, object-orientedness and transformation, ATallows us to examine interactive innovation as a processrather than a product, as dynamic and flexible as opposedto rigid, and as open rather than closed to interpretation.

In light of the growing mobilisation of work and theincreasing diffusion of mobile devices, this study un-iquely addresses the merger of three key interests forscholars and practitioners. First, it looks at co-construc-tive efforts, aimed at the innovation, not the design, oftechnology. Second, it examines innovation as anactivity, not a product, and third, it questions how userinvolvement and mutual learning is affected by mobilework settings. With this background, the question thatmotivated this research project was: How does theinteraction with mobile work affect the innovating oftechnology?

The method of action researchIn pursuit of this question, the researcher needed tobecome part of the social forum that engaged with theinnovative process. Especially for a study of mobileactivities, which are not really ‘observable’ from adistance, proximity to all research subjects was essential.Further, the activity of innovation needed to take placewithin real, mobile settings. As a methodology thatsupports such intervention, while at the same timecontributing to scientific knowledge, Action Researchwas adopted as the most suitable research method(Baskerville & Pries-Heje, 1999; Baskerville & Myers,2004). In this case, all practitioners gained an insight

into the dynamic aspects of their work through the eyesof a researcher and the researcher’s findings wereenriched through actual active participation in thecomplexities of work; ‘research inform[ed] practice andpractice inform[ed] research synergistically’ (Avison et al.,1999, p. 94).

The research study was grounded in a highly complexinteractive innovation project that involved the manu-facturing company Nalle, who as the primary innovatorhad set out to innovate a mobile RFID technology.Moreover, it involved the security services companyMorrison Patrolling, as a secondary user and innovationpartner that agreed to host a number of highly interactivetrials within the boundaries of its organisation. Lastly, thetrialists were Morrison’s respective mobile security guards,the primary users of the technology (Table 1).

Empirical materials were collected over the course of a12-month project in which the chosen AT lens wasoperationalised through Action Research. The principalresearcher actively engaged in both action and researchactivities with the innovator Nalle, mobile workers asprimary users and Morrison as a secondary user of mobileRFID. Evidence was collected through logs of interviews,audio and video recordings of work activities, with andwithout the new mobile RFID device, drawings of workpractices and server data (Table 2). Through the interac-tion of the participants and the interdependency of theiractivities, AT principles were applied in Action Researchto develop an interactive innovation framework.

The Action Research project took place in 2005, andstrictly adhered to the basic stages of (a) diagnosing,(b) action planning, (c) action taking, (d) evaluatingand (e) specifying learning (Susman & Evered, 1978;Baskerville, 1999), which are discussed in the samesequence throughout the remainder of this paper.

A case of interactive innovation of technology formobile workBefore this particular innovation project, RFID was at avery interesting junction; it had existed for quite sometime as asynchronous or mobile systems. Synchronous,mobile handheld RFID devices had been heralded as a‘killer combination not killer application’ (Roberge, 2004,p. 1). The potentially enormous impact of contactlessinteraction on work encouraged the three differentparties’ involvement with the interactive innovation

Table 1 Interactive innovation participants

Subject Individuals

Innovator: Nalle

mobile solutions

Nalle product innovation, R&D and IT

managers

Innovation partner:

Morrison patrolling

Mainly managers and IT staff, secondary

users of mobile RFID and field data

Trialists: Mobile

workers

Mobile security guards, traffic dispatchers,

primary users of the technology



project. However, each party participated for differentreasons, or in AT terms each party was motivated todevelop mobile RFID by its own real-world contradictionsand problems. The first step in Action Research wastherefore to diagnose these different motives according tothe chosen Activity System framework. Built on thisconceptual understanding of motivations in terms ofobject-orientedness and outcome-transformation foreach participant, an action plan for interactive innova-tion was developed and executed.

Diagnosing the motivation for the interactiveinnovation projectAs introduced earlier, mobile device manufacturers havebeen very successful at tailoring to the needs of privateusers. For the innovator Nalle, the current shift to thecorporate clientele meant developing technologies thatwere not generic, off-the-shelve solutions, but that couldbe embedded within various organisational contexts.

Nalle spent immense time and energy on enabling thetechnological aspects of the mobile RFID system, but thereal innovation of the technology and its businessapplications only began after the functionality of itssynchronous communication and auto-identificationtechnology was achieved. Nalle was motivated to com-pete actively in the corporate sector; however, a contra-diction existed between the experience the companypossessed, based on generic technologies, and the skills itnow required. Nalle’s previous activities were focusedsolely on device manufacturing; the company did nothave the expertise needed to tie its technologies specifi-cally to mobile work. Accordingly, Nalle’s motive for theproject was to gain this knowledge to extend its mobileRFID prototype into an auto-identification device thatwould enable new work practices and information flowsacross changing mobile work contexts. Based on theabovementioned contradiction of experience vs skillsrequired, this meant that the technology needed to begrounded in actual work contexts, co-constructed by

Table 2 Action and research activities

Subject Location/medium Action and research activities

Innovator (Nalle) Various physical locations

Meetings, conference calls, email,

written reports, memos

Action: In meetings and interviews, we discussed the development

of mobile RFID technology, its constraints and affordances, the role

of the primary and secondary users. We worked on developing the

device in accordance with the contradictions that emerged, within

Nalle, but also at Morrison and its mobile workforce.

Research: On a conceptual level, what was the motivation for

involving other participants? How could the interaction between

subjects be facilitated? How could empirical materials be collected

and analysed?

Secondary user

(e.g., Morrison patrolling)

Morrison headquarters and regional

offices

Face-to-face meetings, but also email

and telephone communication

Action: In meetings, the technology and its objective affordances

were introduced, including the fundamental properties of mobile

RFID. We discussed the context of work at Morrison, but also out in

the field, and designed trial applications accordingly. Further, we

discussed the role of the technology for Morrison, potential trial

difficulties, access to organisational data, research sites and how

mobile RFID would be implemented throughout the geographically

dispersed and mobile settings.

Research: How do we trigger feedback from primary users? What

was Morrison’s motivation? What transformation was necessary?

How could both action and research be staged at this company?

Primary users

(e.g., Mobile security

guards)

Regional offices, patrolling vehicles,

public spaces, client sites, restaurants,

etc.

Action: The researcher accompanied security personnel on day and

night shifts to learn about the central activities of these mobile

workers. Later, the researcher introduced the device, provided

training, helped troubleshoot the technology if needed, tested its

functionability and introduced the interaction protocol and fora that

were to be followed throughout the project (contact details, issue

logs, change logs, etc.).

Research: Approximately 200 open-ended, formal and information

interviews, conversations and field observations elicited what

functions the technology should offer, but also how mobile RFID

performed in the trials, how it changed mobile work activities and

how the innovation could be improved. Also, how could these

individuals interact with Nalle and Morrison to help negotiate the

functionality of the actual device, etc? How did they interact?



those who would later use it. An interactive and activity-focused approach was seen as elemental to the success ofthis innovation. The project was further supported by thefact that interactive innovation had no precedence inmobile environments, which motivated the design of ainteraction framework that would lay out how innovator,innovation partner and trialists would exchange infor-mation pertinent to the technology under development.

As an innovation partner, Morrison Patrolling wasmotivated to take part in this project to improve thecompany’s ability to understand and manage mobilework. The diagnostic stage uncovered contradictionswithin the organisation between information soughtfrom the field and information available from theworkers. Morrison’s interests in mobile RFID was furthermotivated by the promise to improve such organisationalinformation flows and to overcome many other non-automatic interaction difficulties previously experiencedbetween mobile workers and their stationary colleagues.Morrison was reluctant to wait until a suitable technol-ogy was available, and hesitant to buy an off-the-shelvesolution that could not be tested thoroughly. Interactiveinnovation promised the development of a technologythat was developed around Morrison’s contexts, thatwould address the contradictions within its currentmobile work activities and that could be improvedthroughout the various iterations of innovation cycles.

Although trialists did not instigate the innovationproject directly, indirectly their actual mobile practicesmotivated it. Their participation, however, was motivatedby their need to improve their professional relationshipto their colleagues and superiors by enhancing theinteraction and information flows with them. Contra-dictions within the mobile work activities emphasised adisjoint between the guards’ central activity (ensuring thesecurity of various premises), and the administrative tasksthat often took most of their time. For example,recording every task via paper-based methods was notan appropriate solution for their demanding mobile workenvironments. Similarly, a guard’s use of his mobilephone as a means of constantly updating others on hislocation and current job tasks interrupted his work andthereby contradicted with his central activity. As a result,

mobile workers became involved in the project with theobjective of actively interacting with their managers andwith Nalle throughout the innovation process in order toimprove mobile working conditions and practices.

Lastly, the researcher was interested in examining theeffect of mobility on interaction and innovation. Thisdesirable outcome was an empirically tested and practice-based understanding of innovation, interaction andmobility of work.

Applying the activity lens to the context of allparticipants in the diagnostic stage not only revealedmany interesting contradictions within each activity, butalso unveiled contradictions between them. Each partywas motivated to overcome its own contradictions, butall three pursued a different object transformation andoutcome (Table 3). The shared motivation that broughtall parties together was their common desire for the finaloutcome of the interactive innovation activity, themobile RFID system. In recognition of the uniquenessof the different perspectives (motivations and objecttransformations), and the complexity of an activity thataims to align them, all parties emphasised that aninteraction forum was needed that also placed emphasison the interactive activity itself. Based on these premisesof AT, a practical framework need to be established thatwould outline an interaction system for the duration ofthe innovation activity, including who would commu-nicate with whom about specific topics related to theproject and how. This request provided the motivationfor developing a methodology that would ensure thecontinued discourse between innovator, innovationpartner and trialists.

Action planning: how should participants interactthroughout the project?During action planning, the researcher conducted a base-line study to set the start parameters for the innovationactivity. This included reflecting on the results of thediagnosis stage, and facilitating the interaction betweenthe different participants. For this purpose, the researcherand practitioners (innovation partner, Nalle and trialists)collaborated in designing a normative framework that

Table 3 Motivation for interactive innovation participants

Participant Transformation of object Desired outcome

Innovator Nalle Exploit proof-of-concept (of mobile RFID) in a

manner useful for mobile work contexts.

A mobile RFID device that responds to the

requirements of mobile work environments.

Innovation partner

Morrison

Improve existing information systems by

embedding technologies throughout mobile work

settings.

Improved organisational information flows at

Morrison involving mobile work(ers).

Trialists mobile security

guards

Reduce effort required to log work activities,

replace paperwork with automatic data capture.

Ability to focus solely on central work activity,

without need to report everything manually.

Researcher Examine interaction and innovation in the context

of technology and mobility of work.

Empirically tested and practice-based

understanding of innovation, interaction and

mobility of work.



would lay out how the interaction protocol for theduration of the innovation project.

Clearly, different sets of knowledge existed among theparticipants, which needed to be communicatedand combined in order for interaction to be fruitfuland innovation to be successful (Von Hippel, 2005).In recognition of the complexity of the parties involved,the acknowledgment of different, exclusive skills andsets of knowledge became an important factor in thedevelopment of technology for mobile work. In ATterms, the contradictions between these knowledgesets lead the interactive innovation activity intoexpansive cycles, in which the existing knowledgeasymmetries can be balanced. This information exchangewould anchor the innovation within the context ofthe mobile work setting and ensure that the technologyunder development was suitable for the organisationalcontext.

The innovator’s expertise was undeniably the develop-ment of mobile technology. In this setting, it was Nalle’sknow-how of building a synchronous, mobile RFIDreader and driving the transfer of RFID data from thereader to the corporate back-end system. From aninnovation perspective, Nalle took a solution-basedapproach by offering a technology aimed at solvingorganisational inefficiencies. The innovator’s knowledge,accordingly, was solution-based.

The innovation partner, Morrison Patrolling, possessed adifferent set of knowledge. As experts within the contextof their organisational setting, managers knew thecapabilities and limits of their existing IS and how thesewere tied into their daily operations. In most settings,these practices involved technological IS and manualprocedures, mobile security guards and office-basedmanagers and traffic dispatchers. In other words, theinnovation partner understood how mobile telephony,for instance, and asynchronous, often paper-basedprocedures existed side-by-side within the confines oftheir respective organisational objectives and contexts. Inorder to take advantage of these new developments fortheir organisational processes, the innovation partneroffered context-based knowledge to the interactive innovationproject.

The trialists, Morrison’s mobile security guards, focusedon the actual work to be completed. The immediacy toeveryday tasks placed them into the remarkable positionthat they were the only individuals who really knew whathappened out in the field. Although their superiors hadaccess to paper-based logs and other asynchronousrecords of their work, these were merely accounts of asubsection of mobile work. Only mobile guards, thefuture primary users, understood their work environmentin all of its finesse, and knew the skills and shortcutsneeded to get the work done. Consequently, trialistspossessed needs-based knowledge.

Next, these knowledge sets needed to be combined in away that would allow individuals to transfer theirrespective knowledge to those who required it, to balance

the information asymmetries that existed between them(Von Hippel, 2005). Three different sets of knowledgeshaped the interactive innovation of technology formobile work, which should interactively inform oneanother, eventually allowing the innovator to developtechnology that is responsive to both primary andsecondary users (Figure 2).

Interaction in Innovation Space A: The interaction in thisparticular space involved the innovator and the trialists.While the former would provide input regarding thetechnological possibilities and the possible affordances ofmobile RFID, the trialists would provide the needs-basedknowledge of work requirements and practical feedbackon the functionability of the technology. This representedtalk about work practices and technology development.

Interaction in Innovation Space B: In the diagnosis stage,both innovation partner and trialists realised that thecorporate understanding of mobile work was removedfrom its reality. The input from trialists was needed tobalance their needs-based understanding of mobile workpractices with the context-based knowledge of organisa-tional requirements. Their interaction would provideinput to the improvement of information flows and workat Morrison. This represented talk about work practices andthe future of mobile work.

Interaction in Innovation Space C: In an effort to remainresponsive to actual work practices, the innovator neededto place strong emphasis on the organisational contextsof the innovation partner. Similarly, for the innovationpartner it was important to learn about the actualproperties and capabilities of mobile RFID so that use-cases could be designed and work practices amended. Itwas elemental that Nalle and Morrison Patrolling closelyinteracted so that the new mobile RFID systems andcorporate legacy systems could be prepared and inte-grated. This represented talk about work contexts and systemsdevelopment.

InnovatorNalle C

D

A B

InnovationPartner

Morrison

TrialistsSecurity Guards

Figure 2 Interactive innovation framework.



Collective Interaction in Innovation Space D: Oncecombined, the knowledge exchanges from individualInnovation Spaces (A–C) include all of the differentperspectives that should inform the innovation of a newtechnology. Solution-based, context-based and needs-based knowledge would coalesce as a united result fromindividual dialogues and negotiations within the respec-tive Innovation Spaces. This does not suggest that allthree parties needed to come together physically inInnovation Space D, but rather that the outcomes ofthe previous Innovation Spaces and the resulting knowl-edge bases needed to be combined and aligned. Throughcontinuing interaction between innovator, innovationpartner and trialists, the interactive innovation activitywould always remain relevant and constantly validate themobile RFID technology under development – in theory.

Action taking: enacting the interactive innovationframeworkAfter diagnosing and planning the interactive innovationproject, in the action taking stage it was time for Nalle andMorrison to execute their respective plans. This stepinvolved the action researcher and managers from Nalleand Morrison intervening with the everyday workpractices at Morrison. The interactive framework andmobile RFID system that were planned and prepared inthe previous stage were now implemented.

All three parties were actively engaged with the newtechnology. At Morrison, many areas of the mobile workenvironments were equipped with RFID tags. Over 2000tags were installed, and each security guard was trainedhow to use the mobile RFID reader throughout his workshift. Training sessions were hosted and manuals weredistributed to patrolmen to ensure that participants wereable to use their new tools. Throughout their subsequentworkdays, these mobile workers read tags attached to anumber of objects (e.g., at gates, doors and windows) andselected among responses from the menu on their RFIDreader. For instance, when a property was checked, asecurity guard read its tag and selected ‘secure’ from thehandset’s menu. This tag event and the status updatewere then automatically and synchronously transmittedto Morrison’s main offices. The auto-identification prop-erties of the mobile RFID system virtually eliminatedmanual logs and worksheets and drastically reduced thetime guards had to spend talking on a mobile phone toreport on their whereabouts. Managers, too, had to spendmuch less time manually locating and coordinating thesecurity guards, reports could be drawn up withinminutes and Morison’s customers could access RFIDevents via extranet sites.

The researcher was active with all three parties, helpingNalle reflect on the events in the field and learn from thecontradictions that emerged, working with Morrison onimproving the information flows and redesigning theactual functionality of the mobile RFID system andaccompanying mobile workers throughout their work.As a result of these expansive cycles, various changes were

made to the actual device, the RFID tag technology andwork practices at Morrison.

One trialist was recorded saying: ‘I no longer have toanswer the phone to respond to the constant questionsfrom the dispatcher. He knows my most recent stop, and Ican concentrate on my job. The other thing that is reallyhelpful is that he can look at the electronic log and findout what happened in the past days. I no longer have tocome in to help him read my writing. Now, when we talkit’s about topics that actually make sense’.

But feedback was not positive all the time, as onerespondent pointed out: ‘At the beginning, finding thetags was a bit of a problem. Different guards installed tagsat different locations on the premises. Although we hadagreed before where the tags would be, it often took timeto find them’. Another trialist reported that ‘tags werevery good when they worked, and it was surprising howquickly we got used to the new system. The trouble waswhen a tag had fallen off, or when the reader malfunc-tioned. This hardly happened, but when it did we neededto figure out what was wrong, replace the tag or rebootthe device, which was a nuisance and did not always fixthe problem’.

In terms of interaction, Nalle and Morrison were inregular communication, via email, face-to-face meetingsand telephone conversations. This interaction resulted inthe improvements of the technology and the organisa-tional information systems (Innovation Space C, above).The interaction of mobile workers with Nalle and withMorrison (in Innovation Spaces A and B), however,proved to be more complicated than anticipated, andprimarily fuelled the evaluation stage of the actionresearch project.

Evaluation of the interactive innovation activityThe mobile work at Morrison provided a suitableenvironment for an analysis and evaluation of interactiveinnovation in action. It allowed the researcher tocompare the participants’ ideal, normative interactionplanned and expressed through the Interactive Innova-tion Framework to how these parties actually interacted,substantiated through the empirical evidence collectedduring the action taking stage.

The original assumption of various knowledge sets wasquickly validated. Participants had no expert knowledgebeyond their respective work domains. Nalle in fact knewnothing about security services, either practically orconceptually, and only provided solution-based know-how. Morrison added the much-needed organisationalcontext and trialists offered their needs-based expertisegrounded in their mobile work. The bird’s eye view of thisarray of activities presented in the Interactive InnovationFramework suggested a well-balanced relationship be-tween the three different parties, where the individualparties would interact through the Innovation Spaces.Together, it would appear, the subjects from each activitywould form a collective of individuals with a commongoal in the trial activities. But could their different sets of



knowledge really be communicated effectively or if theywere too ‘sticky’ (Von Hippel, 2005, p. 8), too hard totransfer from one party to another? How did the tidinessof the normative framework hold up against the messi-ness of mobile work?

By and large, the interaction followed protocol.Individuals knew whom to contact to discuss problems,to ask specific questions or to offer suggestions. Mobileworkers contacted their superiors to talk about misalign-ments between new information flows and the reality oftheir work, and they called Nalle’s support-line foranswers or recommendations specific to the device.Regardless, while the interaction procedures were fol-lowed, in principle, interaction did not occur as regularlyor frequently as desirable or warranted by the contra-dictions that emerged. Mobile workers, above all, did notalways interact with the other parties when theyexperienced problems in the field, or when they hadimprovements to suggest. This was evident when theresearcher accompanied them, indicating that funda-mental problems existed within the innovation activitythat kept participants from interacting. The analysis ofthese contradictions unveiled important lessons forfuture interactive innovation studies.

Specifying learningFrom a research perspective, the events that wereobserved during the action taking stage and outlined inthe evaluation stage were the sources of learning. Theproblems that occurred, or in AT parlance the contra-dictions of the collective activities, illustrated an imbal-ance within and between the activity systems ofinnovator, innovation partner and trialists. Some of thesecontradictions could be resolved, leading to an improve-ment of technology, systems and work practices. Forinstance, electromagnetic interference (EMI) occurredbetween a tag and a container that it was mounted to,which led to the re-engineering and thus the improve-ment of the tag technology. From an innovationperspective, these contradictions were meaningful forthe specific product under development, but they did nottruly advance our understanding of interactive innova-tion as an activity.

Higher level contradictions, on the other hand, canfocus on interaction difficulties beyond the setting of thistrial. More generally, they emphasise if and how theuniqueness of mobility shapes the process of interactiveinnovation and vice-versa, and the expansive cyclesprovide the main contribution to the conceptualisationof interactive innovation.

Contradictions of motivationFrom the outset of the trials, all three participant groupsexpressed collaboration readiness (Olson & Olson, 2000);however, they were each driven by different motives.Nalle was motivated by a need to innovate and developsynchronous mobile RFID as a grounded and practice-driven technology; the innovation partner was motivated

by the need to improve its knowledge and control ofmobile work practices. Mobile workers, on the otherhand, were motivated to participate by a wider array ofneeds, including their own desire to advance their workpractices, the felt animosity towards outdated, paper-based systems and the persistent need to justify theirwork to superiors.

These dissimilar motivations imply that participantshad different predispositions towards the outcome, interms of success or failure, of the trials and the interactiveinnovation activity. In this context, the innovator Nallewas fully aware that failure in the trial activities was apossibility, despite all efforts of negotiating the individualelements. However, even in such an occasion, theevaluation of the trials would still be considered avaluable source of learning for further innovation,research and development. At Morrison, the abovemen-tioned EMI problem occurred when the RFID tags couldnot be mounted onto the steel containers, since theseinterfered with the electromagnetic field of the RFIDcommunication. The tags had to be re-engineered towithstand the shielding of the containers. Although thispresented a problem in the context of the innovationpartner and trialists by affecting both the schedule andthe budget of their operations, the continued interactionfocused on the failing artefact, which enabled Nalle toadvance its solution-focused innovation accordingly.

For Morrison as the innovation partner, on the otherhand, the project was a considerable investment that wasvery clearly aimed at solving real, context-specific needsof their understanding of mobile work. Especially giventhe financial commitment and time investment, the trialhad to succeed not to be seen as an ill-investment andwaste of organisational resources. Success was notmeasured in terms of learning, and failure was mostdefinitely seen as an entirely negative outcome.

Trialists measured the success of the trial against how itcontributed to or infringed on their everyday workpractices. In the event of failure of the innovation, therewere no negative consequences, as trialists had no realinvestment in the project, and mobile workers wouldreturn to their usual work practices and tools. None-theless, the properties of the new technology wouldshape their future work practices, so their involvementand interaction in the innovation process had a definitebearing on their own welfare. In this light, trialists wereplaced in a difficult position. In addition to placing theirown needs and interests first, they had to ‘serve twomasters’. Should a mobile worker focus his attention onhelping Nalle develop a new technology or his employer,the innovation partner, understand mobile work andimprove organisational information systems? Thisproblem set aside, how should mobile workers interactwith the other two parties?

Contradictions of mobility and technologyImportant considerations for this study were the mod-alities of mobile work and their potential impact on the



innovative activity. From this perspective, the geographi-cal distance, distribution and mobility of trialists led tocontradictions unique to mobile interactive innovation.In the Morrison case, trialists moved across largegeographical terrains, some wandered or visited and yetothers travelled (Kristoffersen & Ljundberg, 1998). In thatrespect, they covered different distances from their mainplaces of employment (e.g., Morrison Patrolling’s mainoffices). The importance of these physical, geographicalplaces of employment vis-a-vis their mobile placesof work also introduced interesting contradictions forinteractive innovation.

One security guard commented: ‘If there’s no one I cantalk to about using the device, I will not. Sometimes thereader does not work and I need help, but in other casesthere should be another function added. If someone washere, I would tell them, but later I don’t remember y orI just want to go home after my round’.

In the Interaction Framework, the Innovating Spaces Aand B focus on talk about work practices, mobiletechnology development and the improvement of orga-nisational information flows. However, interacting withmobile workers is more easily said than done. For theinnovator and the innovation partner it was difficult toget in touch with mobile workers, to get a practice-basedunderstanding of the needs-based requirements duringthe action planning stage and throughout action taking.Interestingly, this importance of mobility, distance anddistribution appeared to be valid from both sides, asmobile workers found it difficult, too, to interact withother project participants. The immediate needs of theircentral activity, for example, to check a gate at a specifictime required that they moved from location to location,which stood in sharp contrast to the requirements ofinteractive innovation, for example, to return to theoffice to illustrate the shortcoming of the device or totake time to compose a detailed error log.

Another comment was: ‘Sometimes, it [the device] actsup. I just don’t know what to do; I just restart it and hopeit goes away. I don’t have time to try to fix it; I have mynormal work to do. When I am back at the office I try todescribe what happened, but then I cannot show themand it’s hard to put what happened into words becauseI am no techie’.

In their very essence, mobile technologies are to addresssuch problems of spatial, temporal and contextual distance,and to enable and facilitate interaction from a distance. Inthese interactive innovation trials, mobile technologyadopted a dual role, as technology under developmentand technology in use (Bødker, 1991), it became the focusof innovation at the same time as it was used as a tool in itspursuit. Defined by the very nature of their profession andtheir respective motivation for participating in the inter-active innovation projects, innovator, innovation partnerand trialists treated technology quite differently, and instark contradiction to one another.

As outlined above, the motivation of the innovator wasto create devices in general and for the trial specifically.

Here, Nalle transformed a mobile RFID reader based onthe needs and context-based knowledge obtained in theInnovation Spaces. Contrary to mobile workers, Nalle didnot work with mobile RFID at all. Rather, Nalle’s activityinvolved working on the technology. Using a selection ofother tools, engineers and programmers at Nalle saw theartefact as their immediate focus, the content andoutcome of their work, not as a tool for mobile workers.The Innovation partner, on the other hand, was lessconcerned with the innovation of technologies forgeneral use. Attention was closely placed within thecontext of the organisational work. The technologyunder development, as a result, was to improve theorganisational information flows, to contribute to theoverall access to and understanding of mobile work.Accordingly, the innovation partner desired to workvirtually through mobile computing systems to reducethe distance to the situated, mobile spaces. The trialists’main objective was neither to improve generic devices byworking on them, nor to gain access to others’ work byworking through them; mobile workers plainly workedwith technology in the pursuit of their everyday opera-tions. For trialists, interactive innovation aimed atdeveloping tools to improve their working conditionsand mobile work practices.

Viewed from these three perspectives, technologyunder development received a dramatically differentfocus and level of attention from the various participants.While at first sight mobile RFID appeared to be quitefixed in its ability to perform technical functions andsupport mobile workers, the various roles determined thedegree to which individual parties were able to relate toone another. In other words, while a mobile worker sawand treated mobile RFID as a tool, he was less able torelate to RFID as an outcome (for Nalle’s engineer). By thesame token, mobile workers were unable to relate to thehigher-level involvement of the technology for improvedinformation flows for the innovation partner. Similarly,for Nalle’s engineers and programmers it was difficult toview the technology as a tool through the eyes of amobile worker or as an object for improved informationflows in the organisational contexts. Nalle’s engineers’expertise supported an outcome perspective, not a tool orobject perspective. These object, outcome and tool lenseswere intrinsically incongruent and did not allow partici-pants to interact effectively.

In cases where the different participants desired to viewthe technology through a different lens, to balance theirrespective know-how and support each other’s activities,mobile devices, although synchronous communicationtools, did not always prove to be helpful. On the contrary,at times subjects found that reporting a technological orprocedural problem over the mobile phone was particu-larly difficult and did not help bridge mobility-relatedcontradictions. For instance, mobile workers oftenneeded to call the innovator or innovation partner toreport a limitation of the device or to ask for help. Inmany cases, this required that the device was used to



replicate an error (e.g., read a specific tag), which wasdifficult when the trialist was already navigating andholding his mobile phone. One security guard commen-ted: ‘Sometimes, something goes wrong and I call theNalle helpline. Daniel [from the support team] asks mewhat happened, but in order to redo and describe theerror I need to read a tag with the device – but I can’t dothat while I am on the phone’. This is an interestingre-occurrence of the contradiction between tools andobjects outlined earlier. When a worker experienced aproblem with the RFID device, it moved from being a tool(unconsciously working with it) to being an object of hislabour (consciously working on it). When he tried toreport this problem from a distance, via a mobile phone,he discovered that this, too, degraded from an otherwisecommon tool of his work to an object, leaving himessentially with two objects and no tools – another veryinteresting contradiction of tools based on the inherentmobility of the worker.

Representation and contradictory mediationDuring the interactive innovation project, mobile work-ers used the technology under development and subjec-tively interpreted it within the context of their work.Relaying their experiences to other innovation partici-pants (in the Innovation Spaces) fuelled the iterativeinnovation cycles. But given the mobility of the worker,the distribution of innovation partner and innovator andthe ineffectiveness of mobile technology, this knowledgeexchange rested on the exchange of representations. Theactual device under development was the most obviousrepresentation, which as shown before proved to be ahighly contradictory mediator. Moreover, as a mediator itwas ineffective since only Nalle could manipulate it.

One trialist was recorded stating: ‘Some of the featuresare spot on, but some others could be improved. Somechanges are very small and I know what the mobile RFIDdevice should do to work better for me, but of course Icannot make any changes to it’.

Other representations included drawings, use caseabstractions, flow charts and device menus. Here, theinteraction of Nalle, innovation partner and trialistsrelied on the manipulation of these common objects,from a distance, as mediators of their subject-object-subject relationships, leading to two main contradictionsof representations.

In their communicative roles, participants were tovarying degrees used to express, or externalise, theircognitive understanding through mappings and abstrac-tions of their respective work practices and the trialtechnology. While some participants were engineers whowere used to group work and abstract notations, othersdrove security vehicles for a living and were seldomrequired to describe their work to outsiders. Accordingly,their experience with and ability to compose andcomprehend representations of work activities were quitedifferent. The sheer number of different subject cate-gories (project managers, developers, security guards,

traffic managers and dispatchers) made the notion ofrepresentations quite important and interesting. Torequest a summary of the actions and operations ofsecurity guards turned out to be quite difficult; theirexternalisations (e.g., drawings of maps, verbal summa-ries and demonstrations of their tasks) were oftenincomplete and even plain wrong. Reason for thisincompleteness was the routine fashion with which theycarry out their work; calling their unconscious operationsback into the conscious out of the context of work wasfelt to be very problematic. Consequently, the activity ofinteractive innovation was partially based on representa-tions that were difficult to derive and needed to beverified and validated throughout the trial activity,leading in some cases to a need to re-design use-casesand technological devices.

One security guard noted: ‘It is good that you [theresearcher] are here, since I would not know how to talkto the engineers at Nalle about our work, the technologyand the changes. It’s like they have to imagine what wedo here and we have to try to tell them everything aboutour work to help. The other problem is that we almostspeak two different languages. Sometimes what they saymakes no sense to us. It’s good, too, that we have ourmanagers to talk to, although they’re never where we are’.

This statement illustrates that problems emerged whenthese representations and externalisations were exchanged.In their boundary-crossing role (Star & Griesemer, 1989;Carlile, 2002), they shaped the interaction betweeninnovator, innovation partner and trialists. The subjects’incongruent work contexts led to contradictions ofrepresentations as mediators in the Innovation Spaces.Nalle and the innovation partner were quite able to relateto one another’s notations and models (Innovation SpaceC), including technical mappings of the design elementsof mobile RFID, project management charts, charters andflow charts. On the other hand, the experience of themobile workers had to be connected with the work of theinnovation partner and Nalle’s developers (InnovationSpace A and B), so that future improvements of work-flows, hardware, software, middleware and midlets couldincorporate important fixes, minor corrections andrelevant suggestions. Not only were these representationsdifficult to construct in the first place, especially formobile workers, but also their interpretation was highlydependent on sender and recipient, their respective workknow-how, work contexts and frames of reference.Interpreting even the most accurate descriptions out ofcontext, by subjects with a different knowledge base, ledto numerous misunderstandings in the interactive activ-ity and correspondingly to ill alignments of technologicaldevelopments, systems and work practices.

Contradiction of discretionary and standardisedactivitiesMobile work, although collaborative in nature and part ofcollective activities, is frequently carried out alone.Security guards, for instance, communicated through



mobile technologies and other representations with theirpeers and superiors, but conducted their central workactivity of checking premises and securing buildings, etc.by themselves. The majority of contradictions theycommonly encountered in the field did not requirecollaborative decision-making, and mobile workers usedtheir discretion, their good judgement, to form theirdecisions. This was evident when contradictions emergedthat required a quick decision, which were neverdocumented but were part-and-parcel of being a goodmobile security guard. For instance, if a guard faced atraffic jam, he prioritised between alternate routes andchanged the sequence in which he visited his remainingsites. His familiarity with the mobile work environmentdefined him and his mobile peers as the only true expertsof the mobile work environment, which allowed them toexercise various degrees of discretion in their decision-making in the field. Accordingly, discretionary decisionsregarding the multiple contexts are fundamental tomobile work (Pinelle & Gutwin, 2003; Al-Taitoon,2005), but interactive innovation challenges this discre-tionary work in two ways.

Interaction and knowledge exchange requires thatwork practices are formalised and exchanged. Developingabstract notations of otherwise unconscious operationswas already discussed as problematic, as was the creationand exchange of representations that describe extraor-dinary events in the field. Moreover, mobile workersexpressed ambivalence towards disclosing how theyexercise discretion in the field. While they wantedto help the innovation of the mobile RFID device toadvance their mobile work practices, they did not want todisclose the ‘tricks of their trade’, the shortcuts, theworkarounds and the quick fixes that nobody had everasked about or controlled before. For instance, one guardexpressed: ‘We always pinch 20 min here and there; wework long shifts and need breaks. In the meanwhile, thevehicle is empty – they [managers] don’t like that. It’salways been like that and they have always turned a blindeye if we did not overdo it and got the work done. And wealways did. But now they have proof of where we are andwhen, and they have to act on it. They never did thisbefore and we need to figure out a way around this’.

This testimonial shows how mobile workers useddiscretion to structure their workdays, and how theirsuperiors tolerated this level of uncertainty. The inter-active innovation project now required that such beha-viour was disclosed. The notion of embedding suchdiscretionary actions into the functions offered through anew product was seen as contradictory by both parties.Mobile work activities that are highly discretionary allowa continuous negotiation between mobile workers andsuperiors. Inscribing these activities into new innova-tions would put an end not only to the discretion of themobile worker, but the inherent standardisation alsoprevents his ability to negotiate aspects of his work. Forinstance, the RFID system facilitated insights into mobilework that were previously not possible or required, which

was met with resistance by mobile worker and led to animportant contradiction for the interactive innovation oftechnology for mobile work. Perhaps surprisingly, thiswas also seen as contradictory by their superiors, whonow had explicit data on the work conducted in the field,could no longer allow discretionary actions and had nochoice but to reprimand their employees for behaviourthat was previously tolerated.

ConclusionFor Nalle, the appreciation of Innovation Spaces, com-bined with the respective knowledge sets, was recognisedas invaluable for staging interactive innovation projects.Moreover, the Interactive Innovation Framework hadproven beneficial not only for setting out the normativeinteraction model, but also as an analytical tool foridentifying interaction contradictions. All of the contra-dictions that emerged in this project (summarised inTable 4) directly affected the interaction between in-novator, innovation partner and trialists, and indirectlyshaped the technology under development.

Interactive innovation is different in mobile settings.Innovators of mobile technology, striving to involve theirfuture primary and secondary users, face distinct chal-lenges within mobile settings. Contradictions rooted indissimilar motivations provide a complicated foundationfor interactive innovation activities. The unique condi-tions of mobile work then add novel spatial, temporaland contextual contradictions, which mobile technologyitself might be unable to resolve. In fact, the duality of adevice-in-use and a device-under-development can bedisadvantageous to the interaction between participants.As a result, the knowledge exchange with mobile workersrests on asynchronous representations, including nota-tions of work practices and more abstract descriptions ofworkflows. However, such externalisations are hard todevelop, especially by mobile workers who are not oftenrequired to reflect on their work practices, and difficult toexchange and interpret across the different activities ofinnovator, innovation partner and trialists. The fact thatthe different parties work either on, through or withtechnology adds further complication to interactiveinnovation, as individuals are much less able to relateto each other’s work. Lastly, an important contradictionof discretion reveals how mobile workers might object torevealing their unique work practices in the field.Inscribing their actions and operations into new innova-tions suggests standardising previously flexible anddiscrete work practices. For mobile workers, this inher-ently reduces their future ability to negotiate workpractices and to exercise discretion in the field, acontradiction that strongly influences their motivationto interact and support the innovation of a newtechnology for mobile work.

For manufacturers who desire to develop technologiestied to mobile organisational environments, interactiveinnovation activities are important sources of criticalmobility-specific lessons. Although inviting a multitude



of categorically different parties with different activitiesinto the innovation process adds to its complexity, theinteraction with organisational customers and end userspromises access to solution, context and needs-basedknowledge sets. Through their initial input, in combina-tion with contradictions that emerge during the innova-tion activity, innovators are able to go through expansivecycles to improve the technology under development.However, for the interaction within the innovationactivity it is elemental to recognise that mobile workersare motivated to support the development of thetechnology on the one hand, but on the other handthey might resist the impact innovation has on the futureof their mobile work practices.

This activity-based discussion aims to complementthe predominantly product-focused and diffusion-oriented body of innovation and mobility literature.We anticipate that as the mobilisation of work continuesand the mobile technology industry matures, interactiveprojects will become increasingly imperative andactivity-focused methodologies even more crucial forthe successful innovation, and study, of technologiesfor mobile work. We believe that our findings arerelevant for other innovation projects, and hopethat they will not only help motivate research concen-trating on the innovation, but also the design,adoption and appropriation of technologies for mobilework.

About the author

Jan Kietzmann is an assistant professor in the Faculty ofBusiness at Simon Fraser University in Surrey, BritishColumbia, Canada. His Ph.D. research at the London

School of Economics, University of London, specialisedin the intersection of innovation, technology and themobility of work, with a particular focus on emerging

Table 4 Contradictions of interactive innovation

Contradiction Conflict Impact

Motivation Innovator Success desired, failure acceptable and not

negative (solution-focus)

Interactive innovation is flawed if participants

are motivated to transform the object into a

different outcome and if success and failure of

this transformation are to different degrees

acceptable.

Innovation partner Success imperative, failure not acceptable

(context-focus)

Trialist Failure has no impact, success possibly positive

or negative (needs-focus)

Mobility Innovator and

innovation partner

Interactive innovation demands stopping work,

potentially returning to the office, to report

errors or to refine information flows, etc.

Interactive innovation is flawed if it requires

different degrees of mobility. Trialists’ primary

concern is motivated by the mobility required

by their central work activity, not the mobility of

interactive innovation.

Trialist Mobile work activities demand continued

mobility.

Technology Innovator Working on technology Interaction is flawed if parties are unable to

relate to the central activity of each other (e.g.,

developers of technology do not know what it is

like to work through it, or with it).

Innovation partner Working through technology

Trialist Working with technology

Representation Innovator and

innovation partner

Collaborative work externalised through

abstractions of technology or work practices.

Interaction is flawed if trialists are unable to

express their work practices in abstract terms. In

any event, effective interaction is hindered

through interpretation out of context (solution-

vs context- vs needs-based).

Trialist Mostly isolated, individual work, collaborative

mostly for exchange of explicit field data.

Discretion and

control

Innovator Solution-based activity requires feedback from

the field.

Mobile workers’ motivation is to retain flexibility

and ability to (re)negotiate working conditions.

This opposes interactive innovation, which

eventually supports closedness and standardi-

sation. Information from trialists might there-

fore not support interactive innovation.

Innovation partner Context-based activity requires formalisation of

mobile work practices.

Trialist High degree of discretion is a primary condition

of mobile work.



mobile auto-identification technologies. Kietzmann’scurrent research continues to concentrate on mobile IS,but also investigates the topics of interactive innovation

outside of the mobility domain and IT project manage-ment leading up to the Olympic Games in Vancouver in2010.

ReferencesAGRE PE (2001) Changing places: contexts of awareness in computing.

Human-Computer Interaction 16(2–4), 177–192.AL-TAITOON A (2005) Making sense of mobile ICT-enabled trading in

fast moving financial markets as volatility-control ambivalence: casestudy on the organisation of off-premises foreign exchange at aMiddle-East Bank. In London School of Economics, Information Systems.University of London, London.

ANGELL I and KIETZMANN J (2006) RFID and the end of cash?Communications of the ACM 49(12), 90–96.

AVISON D, LAU F, MYERS M and NIELSEN PA (1999) Action research.Communications of the ACM 42(1), 94–97.

BANNON LJ (1997) Activity theory. In CHI 97 from http://acm.org/sigchi/chi97/proceedings/tutorial/bn.htm.

BASKERVILLE R (1999) Investigating information systems with actionresearch. Communications of the AIS 2(19), 2–31.

BASKERVILLE R and MYERS MD (2004) Making IS research relevant topractice. MIS Quarterly 28(3), 329–335.

BASKERVILLE R and PRIES-HEJE J (1999) Grounded action research: a methodfor understanding IT in practice. Management and InformationTechnology 9, 1–23.

BJERKNES G and BRATTETEIG T (1987) Perspectives on description tools andtechniques in system development. In System Design for HumanDevelopment and Productivity: Participation and Beyond (DOCHERTY P,FUCHS-KITTOWSKI K, KOLM P and MATHIASSEN L, Eds), North-Holland,Amsterdam.

BøDKER S (1991) Through the Interface: A Human Activity Approach to UserInterface Design. Lawrence Erlbaum Associates, Hillsdale, NJ.

BøDKER S and GRøNBÆK K (1991) Cooperative prototyping: users anddesigners in mutual activity. International Journal of Man-MachineStudies 34(3), 453–478.

BOURDIEU P (1977) Outline of a Theory of Practice. Cambridge UniversityPress, Cambridge.

BROWN B and O’HARA K (2002) Place as a practical concern of mobileworkers. Environment and Planning A 11(11), 1–23.

CAIRNCROSS F (1997) The Death of Distance. Harvard Business SchoolPress, Boston, MA.

CARLILE PR (2002) A pragmatic view of knowledge and boundaries:boundary objects in new product development. Organization Science13(2), 442–445.

CASTELLS M (1996) The Rise of the Network Society. Blackwell Publishers,Malden, MA.

CHRISTENSEN CM (1997) The Innovator’s Dilemma: When new Techno-logies Cause Great Firms to Fail. Harvard Business School Press,Boston, MA.

CIBORRA CU (2002) The Labyrinths of Information: Challenging the Wisdomof Systems. Oxford University Press Oxford, United Kingdom.

COHEN WM and LEVINTHAL DA (1990) Absorptive capacity: a newperspective on learning and innovation. Administrative Science Quar-terly 35(1), 128–152.

DAVIS J and DEVINNEY T (1997) The Essence of Corporate Strategy: Theoryfor Modern Decision Making. Allen & Unwin, Australia.

EHN P (1993) Scandinavian design: on participation and skill. InParticipatory Design: Principles and Practices (SCHULER D and NAMIOKA

A, Eds), pp 41–77, Lawrence Erlbaum Associates, Mahwah, NJ.EHN P and KYUNG M (1987) The collective resource approach to

system design. In Computers and Democracy: A ScandinavianChallenge (BJERKNES G, EHN P and KYNG M, Eds), Avebury, Aldershot,England.

ENGESTRoM Y (1987) Learning by Expanding – An Activity Theory Approachto Developmental Research. Orienta-Konsultit Oy, Helsinki.

FLOYD C, MEHL W-M, REISIN F-M, SCHMIDT G and WOLF G (1989) Out ofScandinavia: alternative approaches to software design and systemdevelopment. Human-Computer Interaction 4, 253–350.

FONTANA ER and SøRENSEN C (2005) From idea to blah! understandingmobile services development as interactive innovation. In 2ndCONTECSI: International Conference on Information Systems andTechnology Management (RICCIO EL and FLEURY MTL, Eds), Sao Paulo,Brazil.

FORTUNATI L (2001) The mobile phone: an identity on the move. Personaland Ubiquitous Computing 5(2), 85–98.

FRIEDMAN AL and CORNFORD DS (1992) Computer Systems Development:History, Organization and Implementation. New York, John Wiley &Sons, Chichester.

GIBSON JJ (1977) The theory of affordances. In Perceiving, Acting, andKnowing (SHAW R and BRANSFORD J, Eds), Lawrence Erlbaum Associates,Hillsdale, NJ.

HADDON L, MANTE E, SAPIO B, KOMMONEN K-H, FORTUNATI L and KANT AE(2005) Everyday Innovators: Researching the Role of Users in ShapingICTs. Springer, Berlin.

HANSMANN U, MERCK L, NICKLOUS MS and STOBER T (2003) PervasiveComputing: The Mobile World. Springer Verlag, Heidelberg.

IZADI S, PEDRO C and RODDEN T (2002) The FUSE platform: supportingubiquitous collaboration within diverse mobile environments. Auto-mated Software Engineering 9(2), 167–186.

JONES MLW (1995) Where do you want to go tomorrow? Shapingalternative technological futures via ethnographically informed parti-cipatory design. In School of Communication. Simon Fraser University,Vancouver.

KAKIHARA M and SøRENSEN C (2001) Expanding the ‘mobility’ concept.Siggroup Bulletin 22(3), 33–37.

KAPTELININ V and NARDI BA (1997) Activity theory: basic concepts andapplications. In CHI 97. from http://acm.org/sigchi/chi97/proceedings/tutorial/bn.htm.

KIM S, CHO H, HAHM H, LEE S and LEE MS (2003) Interoperability betweenUMTS and CDMA2000 networks. IEEE Wireless Communications 10(1),22–28.

KLEINROCK L (1996) Nomadicity: anytime, anywhere in a disconnectedworld. Mobile Networks and Applications 1, 351–357.

KRISTOFFERSEN S, HERSTAD J, LJUNGBERG F, LøBERSLI F, SANDBAKKEN JRand THORESEN K (1998) Developing scenarios for mobile CSCW.First Workshop on Human Computer Interaction with Mobile Devices34–47.

KRISTOFFERSEN S and LJUNDBERG F (1998) Mobile informatics: innovation ofIT use in mobile settings. IRIS 21 Workshop Report – SIGCHI Bulletin31(1), 29–34.

KRISTOFFERSEN S and LJUNGBERG F (2000) Mobility: from stationary tomobile work. In Planet Internet (BRAA K, Ed), pp 137–156, Studentlit-teratur, Lund.

KUUTTI K (1996) Activity theory as a potential framework for humancomputer interaction research. In Context and Consciousness: ActivityTheory and Human Computer Interaction (NARDI BA, Ed), 17–34, MITPress, Cambridge and London.

LEONTIEV AN (1974) The problem of activity in psychology. SovietPsychology 13(2), 4–33.

LEONTIEV AN (1978) Activity, Consciousness, and Personality. Prentice-Hall,Englewood Cliffs, NJ.

LILISCHKIS S (2003) More yo-yos, pendulums and nomads: trends ofmobile and multi-location work in the information society. In STARProject Issue Report N. 36.

LING R (1998) She calls, [but] it’s for both of us you know: the use oftraditional fixed and mobile telephony for social networking amongNorwegian parents. In R&D Report 33/98. Kjeller, Norway.

LYYTINEN K, VARSHNEY U, ACKERMAN MS, DAVIS G, AVITAL M, ROBEY D,SAWYER S and SøRENSEN C (2004) Surfing the next wave: designand implementation challenges of ubiquitous computing environ-ments – social and organizational issues in ubiquitous computing.



Communications of the Association for Information Systems 13,697–716.

LYYTINEN K and YOO Y (2002a) Issues and challenges in ubiquitouscomputing. Communications of the ACM 45(12), 6–65.

LYYTINEN K and YOO Y (2002b) The next wave of nomadic computing.Information Systems Research 13(4), 377–388.

NARDI BA WHITTAKER S et al (2001) NetWORKers and their activity inintensional networks. Journal of Computer Supported Collaborative Work11(1–2).

OLSON GM and OLSON JS (2000) Distance matters. Human-ComputerInteraction 15, 139–178.

PEDERSEN PE and LING R (2003) Modifying adoption research for mobileinternet service adoption: cross disciplinary interactions. In HawaiiInternational Conference on System Science. IEEE Computer SocietyPress, Big Island, HI.

PICA D and SøRENSEN C (2004) On mobile technology in context:exploring police work. Journal of Computing and Information Technol-ogy 12(4), 287–295.

PIERRE S (2001) Computing and ubiquitous computing: concepts,technologies and challenges. Telematics and Informatics 18, 109–131.

PILLER F, BERGER C, MOESLEIN K and REICHWALD R (2003) Co-Designing theCustomer Interface: Learning from Exploratory Research. TUM BusinessSchool, Munich.

PINELLE D and GUTWIN C (2003) Designing for loose coupling in mobilegroups. In Proceedings of the 2003 International ACM SIGGROUPConference on Supporting Group Work, pp 75–84, ACM Press, SanibelIsland, Florida, USA.

PLANT S (2001) On the Mobile. Motorola, http://www.motorola.com/mot/doc/0/234_MotDoc.pdf.

PORTER M (2001) Strategy and the internet. Harvard Business Review79(3), 62–78.

PORTER M and MILLAR VE (1985) How information gives you competitiveadvantage. Harvard Business Review 63(4), 2–13.

ROBERGE PA (2004) Killer combination not ‘killer application’ defines RFIDand contactless in ‘05. RFID News and RFID Operations, Publisher,http://www.rfidnews.org/news/2004/12/21/killer-combination-not-killer-application-defines-rfid-and-contactless-in-05/ 1 October 2005.

SCARBROUGH H and SWAN J (2005) Developing the process perspective onnetworked innovation. In Theorizing Process in Organizational Research.Santorini, Greece.

SCHEEPERS R, SCHEEPERS H and NGWENYAMA OK (2006) Contextualinfluences on user satisfaction with mobile computing: findings fromtwo healthcare organizations. European Journal of Information Systems15, 261–268.

SHAH S and TRIPSAS M (2004) When do user-innovators start firms?Towards a theory of user entrepreneurship. In University of IllinoisWorking Paper 04-0106.

SøRENSEN C and PICA D (2005) Tales from the police: mobile technologiesand contexts of work. Information and Organization 15(3),125–149.

STAR SL and GRIESEMER JR (1989) Institutional ecology, ‘translations’ andboundary objects: amateurs and professionals in Berkley’s Museumof Veterinary Zoology, 1907–1939. Social Study of Science 19(3),387–420.

SUSMAN G and EVERED R (1978) An assessment of the scientific merits ofaction research. Administrative Science Quarterly 23(4), 582–603.

SWAN J and NEWELL S (2000) Linking knowledge management andinnovation, In European Conference on Information Systems. (BICHLER Mand MAHRER H, Eds) Vienna.

TIDD J, BESSANT J and PAVITT K (1997) Managing Innovation. Wiley,Chichester.

URRY J (2000) Mobile sociology. The British Journal of Sociology 51(1),185–203.

VON HIPPEL E (2005) Democratizing Innovation. The MIT Press, Cam-bridge, MA.

VYGOTSKY LS (1978) Mind in Society – The Development of HigherPsychological Processes. Harvard University Press, Massachusetts,U.S.A.

WHIPP R and CLARK P (1986) Innovation and the Auto Industry: Product,Process and Work Organization. Francis Pinter, London.

WIBERG M (2001) In between mobile meetings: exploring seamlessongoing interaction support for mobile CSCW. In Informatics. UmeaUniversity, Umea.

WIBERG M and GRoNLUND A (2000) Exploring mobile CSCW. In IRIS 23.Laboratorium for Interaction Technology (SVENSSON L, SNIS U, SC, FH,LINDROTH M, MAGNUSSON M and OSTLUND C, Eds), University ofTrollhattan, Uddevalla.

WIREDU GO and SøRENSEN C (2005) The dynamics of control and mobilecomputing in distributed activities. European Journal of InformationSystems 15, 307–319.

